JP2832024B2 - Code transmission method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a code transmission method, and more particularly, to a code transmission method for performing code transmission by adding a redundant code for error detection or correction constituting a product code.

[Prior Art] Conventionally, in the transmission of digital codes, particularly in magnetic recording and the like, it has been necessary to suppress low-frequency components of the recorded digital codes. For example, there is a method for converting an 8-bit code into 9 bits (8/9 conversion) and using only a code that does not include a low-frequency component. However, this method has a problem that the redundancy of the code is increased, and the amount of data to be transmitted is increased.

Therefore, as a method of not increasing the redundancy, there is, for example, mapping coding. Mapping coding is a method that can be applied only to a code sequence having a high correlation between codes, and suppresses low-frequency components of a code sequence by using this correlation. For example, the input code is differentially encoded, and the code near the 0 level is converted into a code with a small DSV (Digital Sum Value) by utilizing the fact that the differential code is concentrated near the 0 level, so that Low frequency components can be suppressed. For example, a 4/4 mapping coding method for converting a 4-bit differential code into a 4-bit code is known.

However, since this mapping coding utilizes the correlation of a code sequence, the effect of suppressing low-frequency components cannot be obtained for codes having no correlation. For example, it is difficult to suppress low-frequency components of the error detection or correction code by the mapping coding as described above. Therefore, the low-frequency suppression effect is reduced even for the entire code sequence including the information code and the error detection or correction code, and the code error rate at the time of decoding is increased.

FIG. 3 is a diagram showing a configuration example of a data frame forming this kind of code sequence. The information data in the figure is the information code sequence subjected to the mapping coding. Similarly, the error detection or correction code is, for example, Hamming. A check point such as a code and a Reed-Solomon code, that is, a redundant code is shown.

FIG. 4 shows an inner code in a data frame as shown in FIG. 3, a plurality of such data frames arranged vertically, an outer code formed vertically, and a product code as a whole. It is a figure showing a data matrix. In the case of such a configuration, the check codes are two-dimensionally arranged, which is suitable for handling image data and the like.

However, considering that each row (each data frame) is sequentially transmitted in a data matrix as shown in FIG. 4, since there is no data correlation in a portion where an inner code and an outer code are continuously transmitted, mapping is performed. The effect of suppressing low frequency components by encoding cannot be expected.
In particular, in a data frame composed of only the outer code and inner code check points, the check points are continuous for a long time, and in this vicinity, the suppression effect of the low-frequency component of the code sequence is significantly reduced. There was a problem that it would.

On the other hand, the applicant of the present application has filed Japanese Patent Application No.
No. 9420 discloses a technique for suppressing the low-frequency components of a code string by distributing test points of each data frame in information codes of each data frame as shown in FIG.

Similarly, Japanese Patent Application No. 60-291172 filed earlier discloses a method of switching a row to be added in each column of this data matrix for an outer code constituting a product code.

[Problems to be Solved by the Invention] However, only the above-described method of distributing the check points in the information code of each data frame has no effect on the outer code constituting the product code as shown in FIG. No countermeasure can be taken, and there remains a problem that the effect of suppressing the low-frequency component of the code sequence is significantly reduced in a data frame including only the check points of the outer code and the inner code.

In addition, this technique can be solved by switching the row to which the outer code forming the product code is added in each column of the data matrix. However, the method of generating the outer code must be changed for each column. In addition, processing such as changing the coding start point of the outer code is required. In particular, when there are a large number of generated outer code check points, it is necessary to select a pattern in which the outer code check points do not continue in the row direction, which requires considerably complicated processing.

The present invention has been made in view of such a problem, and when transmitting an information code by adding an error detection or correction redundant code constituting a product code, it is possible to effectively reduce low frequency components without complicated processing. It is an object of the present invention to provide a code transmission method capable of suppressing the above.

[Means for Solving the Problems] Under such a purpose, according to the present invention, in a code transmission method for adding an error detection or correction redundant code to an information code for transmission, a matrix in which information codes are two-dimensionally arranged The information code is extracted in the upper oblique direction to form a first error detection or correction code, and is added in the row direction of the matrix to form a first error detection or correction code.
After extracting the code of each row of the matrix to which the redundant code for error detection or correction has been added to form a second error detection or correction code and adding it in the row direction of the matrix, the second error detection or correction is performed. The code of each row of the matrix to which the redundant code is added is sequentially transmitted.

Further, as a preferred embodiment of the present invention, the first redundant code for error detection or correction and the second redundant code for error detection or correction are replaced with a second redundant code for error detection or correction. They are distributed and arranged in the information code of each row.

[Operation] According to the code transmission method of the present invention as described above, an information code is extracted in a diagonal direction on a matrix in which information codes are two-dimensionally arranged to generate a first error detection or correction redundant code,
Further, since the code of each row of the matrix to which the first error detection or correction redundant code is added is extracted to generate the second error detection or correction redundant code, the first code constituting the product code is generated.
The information code can be added in the row direction of the matrix in which the two-dimensionally arranged error detection or correction redundant code and the second error detection or correction redundancy code are used, and each row has an error detection or correction redundancy code. The low frequency component can be suppressed by sequentially transmitting the code of each row of the matrix to which the second error detection or correction redundant code is added, instead of being composed of only the code.

Further, according to the above method, as shown in the preferred embodiment of the present invention, the first redundant code for error detection or correction and the second redundant code for error detection or correction are replaced with the second redundant code for error detection or correction. Since it is easy to disperse and arrange in the information code of each row of the matrix to which the correction redundant code is added, a transmission code with a higher low-frequency suppression effect can be obtained by simple processing. .

Example An example of the present invention will be described below.

FIG. 2 is a block diagram showing a schematic configuration of a digital recording device which is a code transmission device as one embodiment of the present invention. In the figure, reference numeral 1 denotes a mapping encoding circuit, which is a modulation circuit that performs low-frequency component suppression using the correlation of digital information input as described above. Reference numeral 2 denotes a memory for expanding the information codes via the mapping encoding circuit into a two-dimensional array, and reference numeral 3 denotes an address control circuit for controlling the addresses of the memory 2.

FIG. 6 shows a data matrix in which synchronous data is added to the information codes arranged two-dimensionally in the memory 2. The arrows in the figure indicate the code transmission order. For this data matrix, in the device of the present embodiment, a case is considered in which an error detection or correction code for detecting or correcting an error occurring on the transmission path has a chain code configuration of an inner code and an outer code. In FIG. 2, 4 is the outer coding circuit, 5 is the inner coding circuit, 6 is a synchronization signal adding circuit, and 7 is a recording unit.

Here, the operation of the outer code encoding circuit 4 will be described. In the transmission apparatus of this embodiment, the information code (data) is extracted in a direction oblique to the transmission direction, that is, the direction of the data frame (row) of the data matrix in FIG. 6 in order to generate the code. It is assumed that the information code is scanned and extracted as shown by the arrow in FIG. The sixth
The synchronization data shown in FIG. 1 and FIG. 1 (A) is added by the synchronization addition circuit 6 at the subsequent stage, but is shown in order to clarify the transmission direction.

When an outer code (first error detection or correction code) is generated by a general method using the information code scanned and extracted as described above, each data frame (row) is generated as shown in FIG. Is added at the end of the data frame in the direction (row direction) of the data frame.

Hereinafter, this is represented by the irreducible polynomial (X ⁴ +) on the Galois field GF (2).
X + 1) will be described as a generator polynomial using a (11,7) shortened cyclic Hamming code. Here, the root vector of (X ¹⁵ -1) is (0010). In this case, the parity check matrix H ₁ for error detection and correction of the code word is Becomes When this is divided into a (4 × 4) unit matrix I ₄ and another part (4 × 7) matrix P, it can be expressed as H ₁ = P ₁ I ₄ (2).

At this time, the generator matrix G ₁ of the (11,7) shortened cyclic hamming code is G ₁ H ₁ ^T = 0 (3) (T at the right shoulder of H ₁ is a symbol representing the transposition of the matrix. ) Becomes The G the data frame structure as shown in Figure 1 is added code for detection and error correction code words (B) by ₁ matrix. Incidentally, the partial matrix I ₄ of an H ₁ matrix corresponds to the inspection point of the code word, the partial matrix I ₇ in G ₁ matrix corresponds to the information point of the code word.

Simply by generating the outer code in such above generator matrix G _1, the outer code inspection point to be uniformly distributed to each data frame on the matrix instead of being concentrated in one data frame (line) Although it is possible, in the present embodiment, the check points of the outer code are further dispersedly arranged in each data frame.

That is, in this embodiment, the outer code check points are distributed and arranged in the information code extracted for generating the outer code. This is because, instead of extracting information codes for each column (vertically) on a data matrix in order to generate an outer code that constitutes a product code as in the prior art, the information codes are obliquely taken, thereby providing a great effect as described later. What happens. Hereinafter, a specific method of distributing the outer code check points in the extracted information code will be described.

First, the check matrix H ₁ corresponds to the leftmost of the code word on the MSB side is a matrix. Accordingly, the fifth to such frame structure diagram for configuring the code, dispersed in a test matrix H ₁ part a unit matrix as comprised in each column is the identity matrix corresponding to the check points of FIG. 5 Consider the new matrix H ₁ ′ that has been placed. This matrix
H ₁ 'are obtained if Hodokose basic row operations of the matrix to the matrix H _1, the resulting matrix H _1' of course is on the coding scheme the same coding scheme and the G ^1.

In the case of the (11,7) shortened cyclic Hamming code as an example, consider distributing check points at 1, 4, 7, and 10 bits from the left of a codeword. FIG. 1C shows the structure of the code word in this case.

At this time, the matrix H ₁ ′ The basic row operation is added to the matrix H ₁ ′ until Get. The _first , fourth, seventh and tenth columns of the matrix H ₁ ′ are combined to form a unit matrix I ₄ .

Next, the matrix in the same manner as the matrix H ₁ to obtain a matrix G ₁
A matrix G ₁ ′ is created from H ₁ ′. The unit matrix I ₄ in the equation (2) represents the _first , fourth, seventh and tenth columns of the matrix H ₁ ′, and P ₁ represents the second,
It is a (4 × 7) matrix composed of 3, 5, 6, 8, 9, and 11 columns.

As described above, the unit matrix I ₄ distributed as described above is converted to I ₇ in the state of being dispersed and the divided matrix P ¹ is also transposed as it is to form P ₁ ^T and match this When, Get. Here as with matrix G _1, it divided partial matrix I ₇ (No. 2,3,5,6,8,9,11 column) in the matrix G ₁ 'corresponds to the information point of the code word.

If the code word is generated by this matrix G ₁ ′, the obtained code word has the configuration shown in FIG. 1C, and the check points for error detection and correction are distributed. Further, the check matrix for the code word matrix H ₁ ', using either a matrix H ₁ can perform error detection and correction, the codeword generated by the matrix G ₁ also, the matrix G _1' in The codewords generated are also on the same coding scheme.

The outer code check points thus generated are arranged in a two-dimensional array on the memory 2 as shown in FIG. 1 (D). However, each outer code is formed obliquely. By scanning the data matrix to which the outer code is added in the horizontal direction (for each row) and sending it to the next stage, the outer code check points are dispersed on the time axis.

Next, a similar operation is performed on the inner code. The data string including the outer code obtained by scanning the data matrix shown in FIG. 1D in the horizontal direction is converted into the input data of the inner code encoding circuit. Treat as Following the distributed arrangement of the outer code check points, consider the generation and the distributed arrangement of inner code check points.

Since the outer codeword is used as the information code of the inner code, the inner code may be a (15,11) cyclic Hamming code. Similarly, a conventional parity check matrix H ₂ in which (X ⁴ + X + 1) is a generator polynomial is Becomes Generator matrix G ₂ for the matrix H ₂ in the same manner as the outer code, Becomes However, also an inner code inspection point in the case of using the matrix G ₂ is concentrated in the second half of the code word. In order to further improve this point and further suppress the occurrence of low-frequency components caused by the check points, in the present embodiment, the inner code check points are not continuous with the outer code check points generated earlier, and the first code check points are not consecutive. The code is distributed and arranged in the code of each row of the matrix in FIG. In other words, consider the configuration of the code word as shown in FIG. 1E such that the inner code check point is arranged between two adjacent adjacent outer code check points. Figure P ₁₁ to P ₁₄ represents the outer code check points added to the previously, P ₂₁ to P ₂₄ represents a code check points among the addition time. The D ₁ to D ₇ are information symbols.

As shown in FIG. 1 (E), a check matrix H ₂ ′ corresponding to a code in which check points are dispersedly arranged is The basic row operation is performed on the matrix H ₂ (Equation (7)) until the form becomes Is represented as Here, the third, seventh, eleventh, and fifteenth columns are unit matrices I
Represents ₄ . As in the case of the outer code, this matrix H ₂ ′
From the matrix G ₂ ′, Get. Here division of 'Similarly matrix G ₂ and the' in the matrix G ₁ submatrix I ₁₁ (first 2,4,6,8,10,12,14 column) corresponds to the information point of the code word. Further, the first, fifth, ninth, and thirteenth columns correspond to outer code check points. _{2 of} the code word generated using this matrix G ₂ ′
The dimensional arrangement is shown in FIG. 1 (F). In this figure, it is understood that both the outer code check point and the inner code check point are dispersed on the time axis by scanning in the horizontal direction (row direction) and sending out to the transmission path.

The code string in which the outer code check points and the inner code check points are dispersed on the time axis as described above is sent to the recording unit 7 and recorded on the recording medium. This makes it very unlikely that a code error will occur during reproduction.

In the code transmission apparatus according to the above-described embodiment, the low-frequency component of the transmitted (recorded) code sequence is extremely small, and the probability that an error occurs in the transmitted code sequence can be extremely reduced. In terms of processing, the process of generating an error detection or correction code is the same as that of the related art, and no additional processing is performed.

In the above-described embodiment, for simplicity of explanation, (1
The description has been made using the shortened cyclic Hamming code and the (15,11) cyclic Hamming code. Therefore, the final code sequence is a code sequence in which the check points and the information code are simply transmitted alternately one bit at a time. Of course, using a multiple code that handles the information code in units of a plurality of bits in words. By performing the same operation, the inspection points can be distributed and arranged.

In addition, by adopting a configuration in which the code length is longer and the check points are dispersedly arranged one word at a time for a plurality of words, error detection can be performed without substantially impairing the effect of the low frequency suppression modulation method applied to the information code. Alternatively, a correction code can be added.

[Effects of the Invention] As described above, according to the code transmission method of the present invention, when an information code is transmitted by adding an error detection or correction redundant code constituting a product code, complicated processing is involved. The low-frequency component can be effectively suppressed without the need.

[Brief description of the drawings]

1 (A) to 1 (F) are diagrams for explaining a code process in a code transmission apparatus as one embodiment of the present invention, and FIG. 1 (A) is a diagram showing a matrix in which information codes are two-dimensionally arranged. FIG. 1B is a diagram showing a code extraction state when an outer code (first error detection or correction redundant code) is generated. FIG. 1B is a diagram showing a data matrix to which an outer code is added by a normal generation matrix; FIG. 1 (C) is a diagram showing a code array after adding an outer code according to the present embodiment for a codeword, FIG. 1 (D) is a diagram showing a data matrix after adding an outer code according to the present embodiment, FIG. FIG. 1E is a diagram showing a code array after the addition of an inner code (second error detection or correction redundant code) according to the present embodiment for a codeword. FIG. 1F is a diagram showing data after the addition of an inner code according to the present embodiment. FIG. 3 is a diagram showing a matrix. FIG. 2 is a block diagram showing a schematic configuration of a code transmission apparatus as one embodiment of the present invention, FIG. 3 is a diagram showing a general data frame configuration, and FIG. 4 is an error detection or correction of a product code configuration. FIG. 5 is a diagram showing a conventional general data matrix to which redundant codes are added, FIG. 5 is a diagram showing a state where test points are dispersedly arranged in information codes, and FIG. 6 is a matrix on which information codes are two-dimensionally arranged. FIG. 3 is a diagram illustrating a scanning direction during transmission. In the figure, 1 is a mapping encoding circuit, 2 is a memory, 3 is an address control circuit, 4 is an outer encoding circuit, 5 is an inner encoding circuit, 6 is a synchronization addition circuit, and 7 is a recording unit.

Claims (4)

(57) [Claims] An information transmission method for transmitting an information code by adding a redundant code for error detection or correction to the information code and transmitting the information code in a diagonal direction on a matrix in which the information codes are two-dimensionally arranged. One redundant code for error detection or correction is formed and added in the row direction of the matrix, and the code of each row of the matrix to which the first redundant code for error detection or correction is added is extracted in the row direction and After forming two error detection or correction redundant codes and adding them in the row direction of the matrix, the codes of each row of the matrix to which the second error detection or correction redundant code is added are sequentially transmitted in the row direction. A code transmission method characterized by the above-mentioned. 2. The method according to claim 1, wherein the first redundant code for error detection or correction and the second redundant code for error detection or correction are stored in the second redundant code.
3. The code transmission method according to claim 1, wherein the code is distributed in the information code of each row of the matrix to which the redundant code for error detection or correction is added. 3. The first redundant code for error detection or correction is extracted in the diagonal direction by distributing unit matrices in a parity check matrix when generating the redundant code for error detection or correction. The code transmission method according to claim 2, wherein the code transmission method is arranged in an information code in a distributed manner. 4. A method for generating a second redundant code for error detection or correction by distributing unit matrices in a parity check matrix at the time of generating the second redundant code for error detection or correction, thereby distributing the second redundant code for error detection or correction to the first redundant code.
The code transmission method according to claim 2, wherein the information is extracted and distributed in an information code extracted in an oblique direction in the code of each row of the matrix to which the redundant code for error detection or correction is added.